Intraosseous versus intravenous access in adult out-of-hospital cardiac arrest: a Bayesian secondary analysis of the PARAMEDIC-3 randomised controlled trial.
Keith Couper, Chen Ji, Charles D Deakin et al.
✦ AI-curated · Sources linked
In 30 seconds
This pre-planned Bayesian secondary analysis of the PARAMEDIC-3 randomised controlled trial compared intraosseous-first with intravenous-first vascular access in adults with out-of-hospital cardiac arrest requiring access. Among 6082 randomised participants, the posterior probability that intraosseous-first improved 30-day survival was 18.4%; the probability it was inferior for any effect was 81.6%. Findings for favourable neurological outcome were comparable, while return of spontaneous circulation was more likely to be lower with intraosseous-first. The analysis used non-informative priors and adjusted for key baseline characteristics.
Key findings
- The posterior probability that intraosseous-first improved 30-day survival (OR > 1.0) was 18.4%.
- The probability of inferior 30-day survival with intraosseous-first was 81.6% for any effect, 55.0% for a small effect (OR < 0.9), and 13.4% for a moderate effect (OR < 0.75).
- For return of spontaneous circulation, the posterior probability that intraosseous-first was inferior (OR < 1.0) was 98.0%.
- Findings for favourable neurological outcome at hospital discharge were comparable to those for 30-day survival.
Why it matters
Vascular access strategy is part of resuscitation care, and these results suggest that an intraosseous-first approach is more likely to be harmful than beneficial for key outcomes in adult out-of-hospital cardiac arrest. The Bayesian estimates also indicate that a moderate or large overall treatment effect is unlikely, helping frame the uncertainty without implying a definitive difference.
What to keep in mind
This is a secondary analysis of a randomised trial, and its probability estimates depend on the specified Bayesian model and non-informative priors.